Lema joins Anthropic's expanded Cyber Verification Program for Claude Mythos 5.1

WHAT HAPPENED

Lema joined Anthropic's expanded Cyber Verification Program, gaining access to Claude Mythos 5.1 with safeguards configured for authorized security research. The program gives verified security teams access to the AI model with adjusted protections that enable security-focused work.

WHO IS INVOLVED

Lema AI and Anthropic partnered on this initiative. Lema is a vendor risk management company that identifies security weaknesses in third-party software. Anthropic built Claude Mythos 5.1, the large language model offered through the Cyber Verification Program.

HOW IT WORKS

Lema uses Claude Mythos 5.1 to trace how third-party software vulnerabilities become real exposure in customer environments. The AI model helps researchers understand attack paths with more depth and speed than previous methods. Lema combines this capability with its existing approach of demonstrating what vendors can actually reach in a customer's environment and delivering fixes customers can apply independently.

WHY IT MATTERS

The integration amplifies Lema's core function of thinking like attackers who exploit vendor relationships to breach customer networks. Access to Mythos 5.1 lets security teams conduct third-party risk assessments more effectively by understanding vulnerability chains at greater scale. This helps organizations move beyond simply asking vendors whether security controls exist to actively mapping and remediating vendor-related exposures.

THE DEAL

Anthropic expanded its Cyber Verification Program to include Lema, enabling the security team to access Claude Mythos 5.1 with custom safeguards for authorized security research.

Read more

• linkedin.com ↗